Searcharxiv⌕ Search

arXiv subjects

Ke Wang

Publications and source records attributed to Ke Wang.

At least 433 records · Page 24Linked to original sources
arXiv2023

Full tunability and quantum coherent dynamics of a driven multilevel system

Tunability of an artificial quantum system is crucial to its capability to process quantum information. However, tunability usually poses significant demand on the design and fabrication of a device. In this work, we demonstrate that Floquet engineering based on longitudinal driving provides distinct possibilities in enhancing the tunability of a quantum system without needing additional resources. In particular, we study a multilevel model based on gate-defined double quantum dots, where coherent interference occurs when the system is driven longitudinally. We develop an effective model to describe the driven dynamics of this multilevel system, and show that it is highly tunable via the driving field. We then illustrate the versatility and rich physics of a driven multilevel system by exploring phenomena such as driving modulation of resonances, adiabatic state transfer, and dark state. In the context of qubit control, we propose noise-resistant quantum gates based on adiabatic passage. The theoretical consideration we present here is rather general, and is in principle valid for other multilevel quantum systems.

cond-mat.mes-hall↗
arXiv2023

Augment and Criticize: Exploring Informative Samples for Semi-Supervised Monocular 3D Object Detection

In this paper, we improve the challenging monocular 3D object detection problem with a general semi-supervised framework. Specifically, having observed that the bottleneck of this task lies in lacking reliable and informative samples to train the detector, we introduce a novel, simple, yet effective `Augment and Criticize' framework that explores abundant informative samples from unlabeled data for learning more robust detection models. In the `Augment' stage, we present the Augmentation-based Prediction aGgregation (APG), which aggregates detections from various automatically learned augmented views to improve the robustness of pseudo label generation. Since not all pseudo labels from APG are beneficially informative, the subsequent `Criticize' phase is presented. In particular, we introduce the Critical Retraining Strategy (CRS) that, unlike simply filtering pseudo labels using a fixed threshold (e.g., classification score) as in 2D semi-supervised tasks, leverages a learnable network to evaluate the contribution of unlabeled images at different training timestamps. This way, the noisy samples prohibitive to model evolution could be effectively suppressed. To validate our framework, we apply it to MonoDLE and MonoFlex. The two new detectors, dubbed 3DSeMo_DLE and 3DSeMo_FLEX, achieve state-of-the-art results with remarkable improvements for over 3.5% AP_3D/BEV (Easy) on KITTI, showing its effectiveness and generality. Code and models will be released.

cs.CV↗
arXiv2023

Janus monolayer ScXY (X$\neq$Y=Cl, Br and I) for piezoelectric and valleytronic application: a first-principle prediction

Coexistence of ferromagnetism, piezoelectricity and valley in two-dimensional (2D) materials is crucial to advance multifunctional electronic technologies. Here, Janus ScXY (X$\neq$Y=Cl, Br and I) monolayers are predicted to be in-plane piezoelectric ferromagnetic (FM) semiconductors with dynamical, mechanical and thermal stabilities. The predicted piezoelectric strain coefficients $d_{11}$ and $d_{31}$ (absolute values) are higher than ones of most 2D materials. Moreover, the $d_{31}$ (absolute value) of ScClI reaches up to 1.14 pm/V, which is highly desirable for ultrathin piezoelectric device application. To obtain spontaneous valley polarization, charge doping are explored to tune the direction of magnetization of ScXY. By appropriate hole doping, their easy magnetization axis can change from in-plane to out-of-plane, resulting in spontaneous valley polarization. Taking ScBrI with 0.20 holes per f.u. as a example, under the action of an in-plane electric field, the hole carriers of K valley turn towards one edge of the sample, which will produce anomalous valley Hall effect (AVHE), and the hole carriers of $Γ$ valley move in a straight line. These findings could pave the way for designing piezoelectric and valleytronic devices.

cond-mat.mtrl-sci↗
arXiv2023

Demystifying What Code Summarization Models Learned

Study patterns that models have learned has long been a focus of pattern recognition research. Explaining what patterns are discovered from training data, and how patterns are generalized to unseen data are instrumental to understanding and advancing the pattern recognition methods. Unfortunately, the vast majority of the application domains deal with continuous data (i.e. statistical in nature) out of which extracted patterns can not be formally defined. For example, in image classification, there does not exist a principle definition for a label of cat or dog. Even in natural language, the meaning of a word can vary with the context it is surrounded by. Unlike the aforementioned data format, programs are a unique data structure with a well-defined syntax and semantics, which creates a golden opportunity to formalize what models have learned from source code. This paper presents the first formal definition of patterns discovered by code summarization models (i.e. models that predict the name of a method given its body), and gives a sound algorithm to infer a context-free grammar (CFG) that formally describes the learned patterns. We realize our approach in PATIC which produces CFGs for summarizing the patterns discovered by code summarization models. In particular, we pick two prominent instances, code2vec and code2seq, to evaluate PATIC. PATIC shows that the patterns extracted by each model are heavily restricted to local, and syntactic code structures with little to none semantic implication. Based on these findings, we present two example uses of the formal definition of patterns: a new method for evaluating the robustness and a new technique for improving the accuracy of code summarization models. Our work opens up this exciting, new direction of studying what models have learned from source code.

cs.PL↗
arXiv2023

Semi-supervised Parametric Real-world Image Harmonization

Learning-based image harmonization techniques are usually trained to undo synthetic random global transformations applied to a masked foreground in a single ground truth photo. This simulated data does not model many of the important appearance mismatches (illumination, object boundaries, etc.) between foreground and background in real composites, leading to models that do not generalize well and cannot model complex local changes. We propose a new semi-supervised training strategy that addresses this problem and lets us learn complex local appearance harmonization from unpaired real composites, where foreground and background come from different images. Our model is fully parametric. It uses RGB curves to correct the global colors and tone and a shading map to model local variations. Our method outperforms previous work on established benchmarks and real composites, as shown in a user study, and processes high-resolution images interactively.

cs.CV↗
arXiv2023

When and Where to Step: Terrain-Aware Real-Time Footstep Location and Timing Optimization for Bipedal Robots

Online footstep planning is essential for bipedal walking robots, allowing them to walk in the presence of disturbances and sensory noise. Most of the literature on the topic has focused on optimizing the footstep placement while keeping the step timing constant. In this work, we introduce a footstep planner capable of optimizing footstep placement and step time online. The proposed planner, consisting of an Interior Point Optimizer (IPOPT) and an optimizer based on Augmented Lagrangian (AL) method with analytical gradient descent, solves the full dynamics of the Linear Inverted Pendulum (LIP) model in real time to optimize for footstep location as well as step timing at the rate of 200~Hz. We show that such asynchronous real-time optimization with the AL method (ARTO-AL) provides the required robustness and speed for successful online footstep planning. Furthermore, ARTO-AL can be extended to plan footsteps in 3D, allowing terrain-aware footstep planning on uneven terrains. Compared to an algorithm with no footstep time adaptation, our proposed ARTO-AL demonstrates increased stability in simulated walking experiments as it can resist pushes on flat ground and on a $10^{\circ}$ ramp up to 120 N and 100 N respectively. For the video, see https://youtu.be/ABdnvPqCUu4. For code, see https://github.com/WangKeAlchemist/ARTO-AL/tree/master.

cs.RO↗
arXiv2023

Proximity-Induced Superconductivity in Epitaxial Topological Insulator/Graphene/Gallium Heterostructures

The introduction of superconductivity to the Dirac surface states of a topological insulator leads to a topological superconductor, which may support topological quantum computing through Majorana zero modes. The development of a scalable material platform is key to the realization of topological quantum computing. Here we report on the growth and properties of high-quality (Bi,Sb)2Te3/graphene/gallium heterostructures. Our synthetic approach enables atomically sharp layers at both hetero-interfaces, which in turn promotes proximity-induced superconductivity that originates in the gallium film. A lithography-free, van der Waals tunnel junction is developed to perform transport tunneling spectroscopy. We find a robust, proximity-induced superconducting gap formed in the Dirac surface states in 5-10 quintuple-layer (Bi,Sb)2Te3/graphene/gallium heterostructures. The presence of a single Abrikosov vortex, where the Majorana zero modes are expected to reside, manifests in discrete conductance changes. The present material platform opens up opportunities for understanding and harnessing the application potential of topological superconductivity.

cond-mat.mes-hall↗
arXiv2023

OAMatcher: An Overlapping Areas-based Network for Accurate Local Feature Matching

Local feature matching is an essential component in many visual applications. In this work, we propose OAMatcher, a Tranformer-based detector-free method that imitates humans behavior to generate dense and accurate matches. Firstly, OAMatcher predicts overlapping areas to promote effective and clean global context aggregation, with the key insight that humans focus on the overlapping areas instead of the entire images after multiple observations when matching keypoints in image pairs. Technically, we first perform global information integration across all keypoints to imitate the humans behavior of observing the entire images at the beginning of feature matching. Then, we propose Overlapping Areas Prediction Module (OAPM) to capture the keypoints in co-visible regions and conduct feature enhancement among them to simulate that humans transit the focus regions from the entire images to overlapping regions, hence realizeing effective information exchange without the interference coming from the keypoints in non overlapping areas. Besides, since humans tend to leverage probability to determine whether the match labels are correct or not, we propose a Match Labels Weight Strategy (MLWS) to generate the coefficients used to appraise the reliability of the ground-truth match labels, while alleviating the influence of measurement noise coming from the data. Moreover, we integrate depth-wise convolution into Tranformer encoder layers to ensure OAMatcher extracts local and global feature representation concurrently. Comprehensive experiments demonstrate that OAMatcher outperforms the state-of-the-art methods on several benchmarks, while exhibiting excellent robustness to extreme appearance variants. The source code is available at https://github.com/DK-HU/OAMatcher.

cs.CV↗
arXiv2023

Dilute Rhenium Doping and its Impact on Intrinsic Defects in MoS2

Substitutionally-doped 2D transition metal dichalcogenides are primed for next-generation device applications such as field effect transistors (FET), sensors, and optoelectronic circuits. In this work, we demonstrate substitutional Rhenium (Re) doping of MoS2 monolayers with controllable concentrations down to 500 parts-per-million (ppm) by metal-organic chemical vapor deposition (MOCVD). Surprisingly, we discover that even trace amounts of Re lead to a reduction in sulfur site defect density by 5-10x. Ab initio models indicate the free-energy of sulfur-vacancy formation is increased along the MoS2 growth-front when Re is introduced, resulting in an improved stoichiometry. Remarkably, defect photoluminescence (PL) commonly seen in as-grown MOCVD MoS2 is suppressed by 6x at 0.05 atomic percent (at.%) Re and completely quenched with 1 at.% Re. Furthermore, Re-MoS2 transistors exhibit up to 8x higher drain current and enhanced mobility compared to undoped MoS2 because of the improved material quality. This work provides important insights on how dopants affect 2D semiconductor growth dynamics, which can lead to improved crystal quality and device performance.

cond-mat.mtrl-sci↗
arXiv2023

Spin vector deviation and the gravitational wave memory effect between two free-falling gyroscopes in the plane wave spacetime

In the plane wave spacetime, we find that there will be a precession angle deviation between two free-falling gyroscopes when gravitational waves passed through. This kind of angle deviation is closely related to the well-known standard velocity memory effect. Initial conditions such as the separation velocity or displacement between the two gyroscopes will affect this angle deviation. The evolutions of the angle deviation are calculated for different cases. We find that in some extreme circumstance, the angle deviation's order of magnitude produced by a rotating compact binary source could be $10^{-14}$ rads. Therefore, this memory effect caused by the gravitational wave is likely to be detected in the future.

gr-qc↗
arXiv2023

Evidence of high-mass star formation through multi-scale mass accretion in hub-filament-system clouds

We present a statistical study of a sample of 17 hub-filament-system (HFS) clouds of high-mass star formation using high-angular resolution ($\sim$1-2 arcsecond) ALMA 1.3mm and 3mm continuum data. The sample includes 8 infrared (IR)-dark and 9 IR-bright types, which correspond to an evolutionary sequence from the IR-dark to IR-bright stage. The central massive clumps and their associated most massive cores are observed to follow a trend of increasing mass ($M$) and mass surface density ($Σ$) with evolution from IR-dark to IR-bright stage. In addition, a mass-segregated cluster of young stellar objects (YSOs) are revealed in both IR-dark and IR-bright HFSs with massive YSOs located in the hub and the population of low-mass YSOs distributed over larger areas. Moreover, outflow feedback in all HFSs are found to escape preferentially through the inter-filamentary diffuse cavities, suggesting that outflows would render a limited effect on the disruption of the HFSs and ongoing high-mass star formation therein. From the above observations, we suggest that high-mass star formation in the HFSs can be described by a multi-scale mass accretion/transfer scenario, from hub-composing filaments through clumps down to cores, that can naturally lead to a mass-segregated cluster of stars.

astro-ph.GA↗
arXiv2023

DeepMatcher: A Deep Transformer-based Network for Robust and Accurate Local Feature Matching

Local feature matching between images remains a challenging task, especially in the presence of significant appearance variations, e.g., extreme viewpoint changes. In this work, we propose DeepMatcher, a deep Transformer-based network built upon our investigation of local feature matching in detector-free methods. The key insight is that local feature matcher with deep layers can capture more human-intuitive and simpler-to-match features. Based on this, we propose a Slimming Transformer (SlimFormer) dedicated for DeepMatcher, which leverages vector-based attention to model relevance among all keypoints and achieves long-range context aggregation in an efficient and effective manner. A relative position encoding is applied to each SlimFormer so as to explicitly disclose relative distance information, further improving the representation of keypoints. A layer-scale strategy is also employed in each SlimFormer to enable the network to assimilate message exchange from the residual block adaptively, thus allowing it to simulate the human behaviour that humans can acquire different matching cues each time they scan an image pair. To facilitate a better adaption of the SlimFormer, we introduce a Feature Transition Module (FTM) to ensure a smooth transition in feature scopes with different receptive fields. By interleaving the self- and cross-SlimFormer multiple times, DeepMatcher can easily establish pixel-wise dense matches at coarse level. Finally, we perceive the match refinement as a combination of classification and regression problems and design Fine Matches Module to predict confidence and offset concurrently, thereby generating robust and accurate matches. Experimentally, we show that DeepMatcher significantly outperforms the state-of-the-art methods on several benchmarks, demonstrating the superior matching capability of DeepMatcher.

cs.CV↗
arXiv2023

ATOMS: ALMA Three-millimeter Observations of Massive Star-forming regions -- XV. Steady Accretion from Global Collapse to Core Feeding in Massive Hub-filament System SDC335

We present ALMA Band-3/7 observations towards "the Heart" of a massive hub-filament system (HFS) SDC335, to investigate its fragmentation and accretion. At a resolution of $\sim0.03$ pc, 3 mm continuum emission resolves two massive dense cores MM1 and MM2, with $383(^{+234}_{-120})$ $M_\odot$ (10-24% mass of "the Heart") and $74(^{+47}_{-24})$ $M_\odot$, respectively. With a resolution down to 0.01 pc, 0.87 mm continuum emission shows MM1 further fragments into six condensations and multi-transition lines of H$_2$CS provide temperature estimation. The relation between separation and mass of condensations at a scale of 0.01 pc favors turbulent Jeans fragmentation where the turbulence seems to be scale-free rather than scale-dependent. We use the H$^{13}$CO$^+$ (1-0) emission line to resolve the complex gas motion inside "the Heart" in position-position-velocity space. We identify four major gas streams connected to large-scale filaments, inheriting the anti-clockwise spiral pattern. Along these streams, gas feeds the central massive core MM1. Assuming an inclination angle of $45(\pm15)^{\circ}$ and a H$^{13}$CO$^+$ abundance of $5(\pm3)\times10^{-11}$, the total mass infall rate is estimated to be $2.40(\pm0.78)\times10^{-3}$ $M_\odot$ yr$^{-1}$, numerically consistent with the accretion rates derived from the clump-scale spherical infall model and the core-scale outflows. The consistency suggests a continuous, near steady-state, and efficient accretion from global collapse, therefore ensuring core feeding. Our comprehensive study of SDC335 showcases the detailed gas kinematics in a prototypical massive infalling clump and calls for further systematic and statistical analyses in a large sample.

astro-ph.GA↗
arXiv2023

ATOMS: ALMA Three-millimeter Observations of Massive Star-forming regions -XIV. Properties of resolved UC Hii regions

Hydrogen recombination lines (RRLs) are one of the major diagnostics of the physical properties of H{\sc ii} regions. We use RRL H40$α$, He40$α$ and 3 mm continuum emission to investigate the properties of a large sample of resolved UC H{\sc ii} regions identified in the ATOMS survey. In total, we identify 94 UC H{\sc ii} regions from H40$α$ emission. The basic parameters for these UC H{\sc ii} regions such as electron density, emission measure, electron temperature, ionic abundance ratio (n$_{\rm He^+}$/n$_{\rm H^+}$), and line width are derived. The median electron density and the median n$_{\rm He^+}$/n$_{\rm H^+}$ ratio of these UC H{\sc ii} regions derived from RRLs are $\sim$9000 cm$^{-3}$ and 0.11, respectively. Within UC H{\sc ii} regions, the n$_{\rm He^+}$/n$_{\rm H^+}$ ratios derived from the intensity ratio of the He40$α$ and H40$α$ lines seems to be higher in the boundary region than in the center. The H40$α$ line width is mainly broadened by thermal motion and microturbulence. The electron temperature of these UC H{\sc ii} regions has a median value of $\sim$6700 K, and its dependence on galactocentric distance is weak.

astro-ph.GA↗
arXiv2023

Fragmentation of the High-mass "Starless'' Core G10.21-0.31: a Coherent Evolutionary Picture for Star Formation

G10.21-0.31 is a 70 $μ$m-dark high-mass starless core ($M>300$ $\mathrm{M_{\odot}}$ within $r<0.15$ pc) identified in $Spitzer$, $Herschel$, and APEX continuum surveys, and is believed to harbor the initial stages of high-mass star formation. We present ALMA and SMA observations to resolve the internal structure of this promising high-mass starless core. Sensitive high-resolution ALMA 1.3 mm dust continuum emission reveals three cores of mass ranging 11-18 $\mathrm{M_{\odot}}$, characterized by a turbulent fragmentation. Core 1, 2, and 3 represent a coherent evolution at three different evolutionary stages, characterized by outflows (CO, SiO), gas temperature ($\mathrm{H_2CO}$), and deuteration ($\mathrm{N_2D^+/N_2H^+}$). We confirm the potential to form high-mass stars in G10.21 and explore the evolution path of high-mass star formation. Yet, no high-mass prestellar core is present in G10.21. This suggests a dynamical star formation where cores grow in mass over time.

astro-ph.GA↗
arXiv2023

Random perturbation of low rank matrices: Improving classical bounds

Matrix perturbation inequalities, such as Weyl's theorem (concerning the singular values) and the Davis-Kahan theorem (concerning the singular vectors), play essential roles in quantitative science; in particular, these bounds have found application in data analysis as well as related areas of engineering and computer science. In many situations, the perturbation is assumed to be random, and the original matrix has certain structural properties (such as having low rank). We show that, in this scenario, classical perturbation results, such as Weyl and Davis-Kahan, can be improved significantly. We believe many of our new bounds are close to optimal and also discuss some applications.

math.NA↗
arXiv2022

Asymptotic Properties of Random Restricted Partitions

We study two types of probability measures on the set of integer partitions of $n$ with at most $m$ parts. The first one chooses the random partition with a chance related to its largest part only. We then obtain the limiting distributions of all of the parts together and that of the largest part as $n$ tends to infinity while $m$ is fixed or tends to infinity. In particular, if $m$ goes to infinity not too fast, the largest part satisfies the central limit theorem. The second measure is very general. It includes the Dirichlet distribution and the uniform distribution as special cases. We derive the asymptotic distributions of the parts jointly by taking limits of $n$ and $m$ in the same manner as that in the first probability measure.

math.PR↗
arXiv2022

ResGrad: Residual Denoising Diffusion Probabilistic Models for Text to Speech

Denoising Diffusion Probabilistic Models (DDPMs) are emerging in text-to-speech (TTS) synthesis because of their strong capability of generating high-fidelity samples. However, their iterative refinement process in high-dimensional data space results in slow inference speed, which restricts their application in real-time systems. Previous works have explored speeding up by minimizing the number of inference steps but at the cost of sample quality. In this work, to improve the inference speed for DDPM-based TTS model while achieving high sample quality, we propose ResGrad, a lightweight diffusion model which learns to refine the output spectrogram of an existing TTS model (e.g., FastSpeech 2) by predicting the residual between the model output and the corresponding ground-truth speech. ResGrad has several advantages: 1) Compare with other acceleration methods for DDPM which need to synthesize speech from scratch, ResGrad reduces the complexity of task by changing the generation target from ground-truth mel-spectrogram to the residual, resulting into a more lightweight model and thus a smaller real-time factor. 2) ResGrad is employed in the inference process of the existing TTS model in a plug-and-play way, without re-training this model. We verify ResGrad on the single-speaker dataset LJSpeech and two more challenging datasets with multiple speakers (LibriTTS) and high sampling rate (VCTK). Experimental results show that in comparison with other speed-up methods of DDPMs: 1) ResGrad achieves better sample quality with the same inference speed measured by real-time factor; 2) with similar speech quality, ResGrad synthesizes speech faster than baseline methods by more than 10 times. Audio samples are available at https://resgrad1.github.io/.

eess.AS↗